Evaluation of a questionnaire used in dermatological epidemiology. Discrepancy between self-reported symptoms and objective signs.

A questionnaire about facial skin problems was answered by 3745 office employees with the aim of discovering whether VDU work causes skin disease. From this group, 809 randomly selected persons were examined and interviewed by a dermatologist. There was a relatively good correspondence (87%) between the results from the interviews and the questionnaires, but the questionnaire results corresponded to current status in only 46% of the cases. No important difference was seen between those who were informed and those who were not informed about the purpose of the investigation. There was a relatively wide variation in the correlation between two responses to the same questions. In conclusion, the self-administered questionnaire seems to give reasonably adequate answers regarding the occurrence of more clear-cut symptoms over a period of time, rather than answers regarding the presence of signs at a particular point in time. However, less marked changes might not be picked up particularly well and the use of questionnaires alone cannot be recommended when studying mild skin problems due to a particular exposure in a population.